- Single Book
3
- 10.1007/978-3-030-52246-9
Intelligent Computing
- Jan 01, 2020
- Kohei Arai + 2 more +2
Publications from 2021 to 2026
Showing 4 of 4 papers
Intelligent Computing
Intelligent Systems and Applications
Neural networks are frequently used for image classification, but can be vulnerable to misclassification caused by adversarial images. Attempts to make neural network image classification more robust have included variations on preprocessing (cropping, applying noise, blurring), adversarial training, and dropout randomization. In this paper, we implemented a model for adversarial detection based on a combination of two of these techniques: dropout randomization with preprocessing applied to images within a given Bayesian uncertainty. We evaluated our model on the MNIST dataset, using adversarial images generated using Fast Gradient Sign Method (FGSM), Jacobian-based Saliency Map Attack (JSMA) and Basic Iterative Method (BIM) attacks. Our model achieved an average adversarial image detection accuracy of 97%, with an average image classification accuracy, after discarding images flagged as adversarial, of 99%. Our average detection accuracy exceeded that of recent papers using similar techniques.
Read moreComparison of the Effectiveness of Mindfulness Education and Emotional Focused Therapy on Couple Intimacy, Commitment and Happiness Women with Couple Burnout
Comparison of the Effectiveness of Mindfulness Education and Emotional Focused Therapy on Couple Intimacy, Commitment and Happiness Women with Couple Burnout
Read moreEndoscopic Esophagitis is More Severe in Gastroesophageal Reflux Patients With a Positive Family History
The role of genetic factors in the etiology of gastroesophageal reflux disease (GERD) is still uncertain. To define whether the presence of reflux symptoms in first-degree relatives can affect the severity of the endoscopic picture of patients with GERD and disease evolution during follow-up. A total of 1930 consecutive patients with GERD were referred for endoscopy from Trikala prefecture, had an entry endoscopy and a follow-up if needed. Before endoscopic evaluation, all patients and their first-degree relatives completed Reflux Symptom Questionnaire. Patients were followed up for 4 years with Reflux Symptom Questionnaire every 6 months. A total of 258 (62.9%) patients with positive and 724 (47.6%) with negative family history of GERD had esophagitis (P<0.0001). Seventy-six (74.5%) patients with more than 1 family member with GERD had esophagitis (P<0.0001). During follow-up endoscopic picture was aggravated in 101 (25%) patients with positive and 46 (3%) with negative family history. A total of 359 (24%) of GERD patients with negative and 24 (10%) with positive family history managed to stop proton pump inhibitors during follow-up (P<0.0001). In logistic regression analysis: age, male sex, presence of hiatal hernia, family history of GERD, tranquilizer use, frequency, and duration of reflux symptoms were independently associated with presence of esophagitis. Although we cannot overlook the importance of confounding factors such as body weight and/or psychological factors, we found that endoscopic picture is more severe among GERD patients with at least 1 first-degree relative with GERD. During follow-up, patients with negative family history had more chances to wean off proton pump inhibitors after life-style modifications.
Read more